Tamperbox MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| resources | {
"listChanged": false
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| searchB | Discover DNS, email authentication/routing, TLS, website security, network diagnostics, site files, JWT verification/signing, timestamp workflows and API conversion. Includes pasted-evidence analysis and builders. Run an async JavaScript function body over the complete local OpenAPI spec and derived operations array. For website categories and related pages, inspect spec["x-tamperbox-directory"]; browser-only and website-only entries have no executable operation IDs. Filter API operations by tags, descriptions and schemas; return only relevant endpoints and examples. Example: return operations.filter(o => /jwt/i.test(o.operationId)); No API requests or private inputs are needed. Limits: 64 KiB code, 15 seconds by default, 2 MiB returned JSON. |
| executeA | Run an async JavaScript function body in an isolated runtime. Search schemas first, then call await tamperbox.request({method:"POST",path:"/v1/tools/json_format",body:{input:{mediaType:"application/json",value:"{}"}}}), combine results and return JSON. method/path must be documented; query is a scalar-value object; body must match the schema. Returns the decoded JSON response body. Inputs go to Tamperbox; target-check tools can contact the requested public target. No arbitrary URLs, headers, files, imports, process or secrets. Limits: 64 KiB code, 20 bridge calls, 4 in flight, 15 seconds by default, 2 MiB output. JWT decoding does not verify trust. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| catalog |
TDQS
Scored across 2 tools
The two tools are mostly distinct: search is for discovering/filtering the OpenAPI spec and examples without making API calls, while execute is for making documented API requests via tamperbox.request. There is still some overlap because both run async JavaScript bodies, and search mentions builders/evidence analysis, which could blur the boundary slightly.
Both names are lowercase imperative verbs, so the style is consistent and readable. They do not follow a descriptive verb_noun pattern, but for a two-tool meta-server the naming is predictable enough.
Two tools is borderline thin, especially given the very broad domain implied by DNS, email, TLS, website, network, JWT, timestamp, and conversion workflows. The search+execute meta-tool pattern can justify a small surface, but it depends heavily on agents understanding the OpenAPI discovery workflow.
Together, search and execute provide a generic discover-then-call surface that can cover all documented operations in the underlying API. Minor gaps remain because there are no dedicated convenience tools or direct operation listings, but the meta-tool model avoids hard dead ends if the spec is complete.